Way to turn off Newest Photos?

[digitalpoint]

This is a relevant snippet from the only new search api method:

$search_core = vB_Search_Core::get_instance();        
$current_user = new vB_Legacy_CurrentUser();
if (!$vbulletin->options['enablesearches']) {            
    return $this->error('searchdisabled');        
}                
$criteria = $search_core->create_criteria(vB_Search_Core::SEARCH_ADVANCED);                
$userids_a = explode(',', $userids);        
$contenttypeids_a = explode(',', $contenttypeids); 

And so on, let me know if that's not enough to judge compatibility Digital.

At first glance it looks okay... as long as the $search_core object is being reused to call methods. After the $search_core object is initiated, there should no longer be any direct calls to the vB_Search_Core class/methods, otherwise you will end up bypassing the defined search instance.

Referencing constant variables within the class like vB_Search_Core::SEARCH_ADVANCED is fine.

There was a similar issue within the normal search.php page where we were initiating the search class, but then going back and referencing methods via vB_Search_Core::...() calls. http://tracker.vbulletin.com/browse/VBIV-8822

Long story short is as long as vB_Search_Core methods are called from the $search_core object, and not from direct vB_Search_Core calls, it should be fine. It's hard to tell/trace without seeing all the code after the $search_core object init, but that would be the first place I would look.

If it's something that isn't blatantly obvious, I could probably find the underlying cause if I had access to a test site that was exhibiting the issue.

[Kyle Furlong]

At first glance it looks okay... as long as the $search_core object is being reused to call methods. After the $search_core object is initiated, there should no longer be any direct calls to the vB_Search_Core class/methods, otherwise you will end up bypassing the defined search instance.

Referencing constant variables within the class like vB_Search_Core::SEARCH_ADVANCED is fine.

There was a similar issue within the normal search.php page where we were initiating the search class, but then going back and referencing methods via vB_Search_Core::...() calls. http://tracker.vbulletin.com/browse/VBIV-8822

Long story short is as long as vB_Search_Core methods are called from the $search_core object, and not from direct vB_Search_Core calls, it should be fine. It's hard to tell/trace without seeing all the code after the $search_core object init, but that would be the first place I would look.

If it's something that isn't blatantly obvious, I could probably find the underlying cause if I had access to a test site that was exhibiting the issue.

So using vB_Search_Results::create_from_cache($current_user, $criteria); will not work with Sphinx?

[tpearl5]

btw, I opened a ticket with support regarding the newest and popular threads not working on monday but haven't received a response.

[digitalpoint]

So using vB_Search_Results::create_from_cache($current_user, $criteria); will not work with Sphinx?

It should be fine... we don't touch anything in the vB_Search_Results class, unless you are searching for something that vB doesn't normally support with search (like searching PMs or User Notes), but I assume the FB App isn't trying to do that.

[tpearl5]

I made a ticket specifically for this in the JIRA. I really hope this isn't responded with 'we don't support modifications' since it is apparently not digitalpoint's error. It sounds like a consistency issue more than anything.

[Kyle Furlong]

I made a ticket specifically for this in the JIRA. I really hope this isn't responded with 'we don't support modifications' since it is apparently not digitalpoint's error. It sounds like a consistency issue more than anything.

Your forum appears to be misconfigured. I am getting HTTP 400 errors when trying to use the API. Please open a support ticket to troubleshoot this.

[tpearl5]

FYI, I installed the FB app on my other site and it appears to be working without a problem. That includes the problems I was having with sphinx.

So whatever the problem is, it's definitely because of a configuration somewhere localized to CPF